Rob Clark [Fri, 30 Jan 2015 14:48:11 +0000 (20:18 +0530)]
drm/msm/mdp4: Support NV12MT format in mdp4
Using fb modifier flag, support NV12MT format in MDP4.
v2:
- rework the modifier's description [Daniel Vetter's comment]
- drop .set_mode_config() callback [Rob Clark's comment]
v3:
- change VENDOR's name and restrict usage to NV12 [pointed by Daniel]
Signed-off-by: Rob Clark <robdclark@gmail.com>
Rob Clark [Wed, 20 May 2015 14:50:30 +0000 (10:50 -0400)]
drm/msm: update generated headers
Signed-off-by: Rob Clark <robdclark@gmail.com>
Rob Clark [Thu, 4 Jun 2015 14:12:22 +0000 (10:12 -0400)]
drm/msm: add missing DRIVER_ATOMIC flag
Somehow this got lost when msm atomic support was first merged.
Signed-off-by: Rob Clark <robdclark@gmail.com>
Rob Clark [Sun, 19 Apr 2015 14:14:09 +0000 (10:14 -0400)]
drm/msm/adreno: dump scratch regs and other info on hang
Dump a bit more info when the GPU hangs, without having hang_debug
enabled (which dumps a *lot* of registers). Also dump the scratch
registers, as they are useful for determining where in the cmdstream
the GPU hung (and they seem always safe to read when GPU has hung).
Note that the freedreno gallium driver emits increasing counter values
to SCRATCH6 (to identify tile #) and SCRATCH7 (to identify draw #), so
these two in particular can be used to "triangulate" where in the
cmdstream the GPU hung.
Signed-off-by: Rob Clark <robdclark@gmail.com>
Dan Carpenter [Wed, 10 Jun 2015 19:27:51 +0000 (22:27 +0300)]
drm/virtio: remove some dead code
The goto is correct, and we never reach the return statement so just
delete the dead code.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dan Carpenter [Wed, 10 Jun 2015 19:27:11 +0000 (22:27 +0300)]
drm/virtio: checking for NULL instead of IS_ERR
virtio_gpu_alloc_object() returns an error pointer, it never returns
NULL.
Fixes: dc5698e80cf7 ('Add virtio gpu driver.')
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Thu, 11 Jun 2015 03:04:07 +0000 (13:04 +1000)]
Merge branch 'drm-next-4.2-amdgpu' of git://people.freedesktop.org/~agd5f/linux into drm-next
More fixes for amdgpu for 4.2. We've integrated Jerome's comments
about the interface among other things. I'll be on vacation next week
so Christian will be handling any updates next week.
* 'drm-next-4.2-amdgpu' of git://people.freedesktop.org/~agd5f/linux: (23 commits)
drm/amdgpu: fix a amdgpu_dpm=0 bug
drm/amdgpu: don't enable/disable display twice on suspend/resume
drm/amdgpu: fix UVD/VCE VM emulation
drm/amdgpu: enable vce powergating
drm/amdgpu/iceland: don't call smu_init on resume
drm/amdgpu/tonga: don't call smu_init on resume
drm/amdgpu/cz: don't call smu_init on resume
drm/amdgpu: update to latest gfx8 golden register settings
drm/amdgpu: whitespace cleanup in gmc8 golden regs
drm/admgpu: move XDMA golden registers to dce code
drm/amdgpu: fix the build on big endian
drm/amdgpu: cleanup UAPI comments
drm/amdgpu: remove AMDGPU_CTX_OP_STATE_RUNNING
drm/amdgpu: remove the VI hardware semaphore in ring sync
drm/amdgpu: set the gfx config properly for all CZ variants (v2)
drm/amdgpu: also print the pci revision when printing the pci ids
drm/amdgpu: cleanup VA IOCTL
drm/amdgpu: fix saddr handling in amdgpu_vm_bo_unmap
drm/amdgpu: fix amdgpu_vm_bo_map
drm/amdgpu: remove unused AMDGPU_IB_FLAG_GDS
...
Sonny Jiang [Wed, 10 Jun 2015 17:46:36 +0000 (13:46 -0400)]
drm/amdgpu: fix a amdgpu_dpm=0 bug
Signed-off-by: Sonny Jiang <sonny.jiang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Wed, 10 Jun 2015 15:52:00 +0000 (11:52 -0400)]
drm/amdgpu: don't enable/disable display twice on suspend/resume
We were doing it in the common code and in the IP specific code.
Remove the IP specific code. The common code handles the
ordering properly.
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Christian König [Wed, 10 Jun 2015 15:20:11 +0000 (17:20 +0200)]
drm/amdgpu: fix UVD/VCE VM emulation
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Sonny Jiang [Thu, 28 May 2015 19:47:53 +0000 (15:47 -0400)]
drm/amdgpu: enable vce powergating
Enable VCE dpm and powergating. VCE dpm dynamically scales the VCE clocks on
demand.
Signed-off-by: Sonny Jiang <sonny.jiang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Alex Deucher [Tue, 9 Jun 2015 21:46:30 +0000 (17:46 -0400)]
drm/amdgpu/iceland: don't call smu_init on resume
smu_init allocates buffers and initializes them. It does not
touch the hw. There is no need to do it again on resume. It
should really be part of sw_init (and smu_fini should be part
of sw_fini), but we need the firmware sizes from the other IPs
for firmware loading so we have to wait until sw init is done
for all other IPs.
Reviewed-by: Sonny Jiang <Sonny.Jiang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Tue, 9 Jun 2015 21:42:10 +0000 (17:42 -0400)]
drm/amdgpu/tonga: don't call smu_init on resume
smu_init allocates buffers and initializes them. It does not
touch the hw. There is no need to do it again on resume. It
should really be part of sw_init (and smu_fini should be part
of sw_fini), but we need the firmware sizes from the other IPs
for firmware loading so we have to wait until sw init is done
for all other IPs.
Reviewed-by: Sonny Jiang <Sonny.Jiang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Tue, 9 Jun 2015 21:32:53 +0000 (17:32 -0400)]
drm/amdgpu/cz: don't call smu_init on resume
smu_init allocates buffers and initializes them. It does not
touch the hw. There is no need to do it again on resume. It
should really be part of sw_init (and smu_fini should be part
of sw_fini), but we need the firmware sizes from the other IPs
for firmware loading so we have to wait until sw init is done
for all other IPs.
Reviewed-by: Sonny Jiang <Sonny.Jiang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Tue, 9 Jun 2015 18:33:03 +0000 (14:33 -0400)]
drm/amdgpu: update to latest gfx8 golden register settings
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Tue, 9 Jun 2015 18:06:18 +0000 (14:06 -0400)]
drm/amdgpu: whitespace cleanup in gmc8 golden regs
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Tue, 9 Jun 2015 17:51:25 +0000 (13:51 -0400)]
drm/admgpu: move XDMA golden registers to dce code
Already moved other display registers.
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Tue, 9 Jun 2015 13:58:23 +0000 (09:58 -0400)]
drm/amdgpu: fix the build on big endian
Some leftover copy and pastes from radeon that never
got updated.
Reviewed-by: Christian König <christian.koenig@amd.com>
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Christian König [Tue, 9 Jun 2015 13:54:37 +0000 (15:54 +0200)]
drm/amdgpu: cleanup UAPI comments
No functional change.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Christian König [Tue, 9 Jun 2015 13:21:35 +0000 (15:21 +0200)]
drm/amdgpu: remove AMDGPU_CTX_OP_STATE_RUNNING
Not used.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
David Zhang [Wed, 10 Jun 2015 18:28:56 +0000 (02:28 +0800)]
drm/amdgpu: remove the VI hardware semaphore in ring sync
Signed-off-by: David Zhang <david1.zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Alex Deucher [Fri, 5 Jun 2015 18:34:19 +0000 (14:34 -0400)]
drm/amdgpu: set the gfx config properly for all CZ variants (v2)
Need to adjust the number of CUs and RBs.
v2: get proper values
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Samuel Li <samuel.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Fri, 5 Jun 2015 19:04:33 +0000 (15:04 -0400)]
drm/amdgpu: also print the pci revision when printing the pci ids
The driver makes use of this information so print if to aid in
debugging.
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Samuel Li <samuel.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Christian König [Mon, 8 Jun 2015 13:03:00 +0000 (15:03 +0200)]
drm/amdgpu: cleanup VA IOCTL
Remove the unnecessary returned status and make the IOCTL write only.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Christian König [Fri, 5 Jun 2015 18:56:17 +0000 (20:56 +0200)]
drm/amdgpu: fix saddr handling in amdgpu_vm_bo_unmap
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Christian König [Fri, 5 Jun 2015 08:06:04 +0000 (10:06 +0200)]
drm/amdgpu: fix amdgpu_vm_bo_map
We need to reset the bo_va address, otherwise new mappings
wouldn't be updated in the page table.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Jammy Zhou [Fri, 5 Jun 2015 20:49:22 +0000 (04:49 +0800)]
drm/amdgpu: remove unused AMDGPU_IB_FLAG_GDS
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Chunming Zhou [Mon, 1 Jun 2015 06:35:03 +0000 (14:35 +0800)]
drm/amdgpu: disable user fence interrupt (v2)
amdgpu submits both kernel and user fences, but just need one interrupt,
disable user fence interrupt and don't effect user fence.
v2: fix merge error
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Christian König [Wed, 3 Jun 2015 19:31:20 +0000 (21:31 +0200)]
drm/amdgpu: fix user ptr race condition
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: monk liu <monk.liu@amd.com>
Jack Xiao [Thu, 4 Jun 2015 04:18:27 +0000 (12:18 +0800)]
drm/amdgpu: add zero timeout check in amdgpu_fence_wait_seq_timeout
Signed-off-by: Jack Xiao <Jack.Xiao@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Dave Airlie [Fri, 5 Jun 2015 02:31:12 +0000 (12:31 +1000)]
virtgpu: include linux/types.h to avoid warning.
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Fri, 5 Jun 2015 01:14:18 +0000 (11:14 +1000)]
Merge branch 'drm-next-4.2-amdgpu' of git://people.freedesktop.org/~agd5f/linux into drm-next
This is the big pull request for amdgpu, the new driver for VI+ AMD
asics. I currently supports Tonga, Iceland, and Carrizo and also
contains a Kconfig option to build support for CI parts for testing.
All major functionality is supported (displays, gfx, compute, dma,
video decode/encode, etc.). Power management is working on Carrizo,
but is still being worked on for Tonga and Iceland.
* 'drm-next-4.2-amdgpu' of git://people.freedesktop.org/~agd5f/linux: (106 commits)
drm/amdgpu: only support IBs in the buffer list (v2)
drm/amdgpu: add vram_type and vram_bit_width for interface query (v2)
drm/amdgpu: add ib_size/start_alignment interface query
drm/amdgpu: add me/ce/pfp_feature_version interface query
drm/amdgpu add ce_ram_size for interface query
drm/amdgpu add max_memory_clock for interface query (v2)
drm/amdgpu: add hdp flush for gfx8 compute ring
drm/amdgpu: fix no hdp flush for compute ring
drm/amdgpu: add HEVC/H.265 UVD support
drm/amdgpu: stop loading firmware with pm.mutex locked
drm/amdgpu: remove mclk_lock
drm/amdgpu: fix description of vm_size module parameter (v2)
drm/amdgpu: remove all sh mem register modification in vm flush
drm/amdgpu: rename GEM_OP_SET_INITIAL_DOMAIN -> GEM_OP_SET_PLACEMENT
drm/amdgpu: fence should be added to shared slot
drm/amdgpu: sync fence of clear_invalids (v2)
drm/amdgpu: max_pde_used usage should be under protect
drm/amdgpu: fix bug of vm_bo_map (v2)
drm/amdgpu: implement the allocation range (v3)
drm/amdgpu: rename amdgpu_ip_funcs to amd_ip_funcs (v2)
...
Frediano Ziglio [Wed, 3 Jun 2015 11:09:19 +0000 (12:09 +0100)]
drm/qxl: Propagate correctly errors from qxlhw_handle_to_bo
This function could return a NULL pointer in case of handle not
present and in case of out of memory conditions however caller
function always returned EINVAL error hiding a possible ENOMEM.
This patch change the function to return the error instead to
be able to propagate the error instead of assuming EINVAL.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Frediano Ziglio [Wed, 3 Jun 2015 11:09:18 +0000 (12:09 +0100)]
drm/qxl: Simplify cleaning qxl processing command
In qxlhw_handle_to_bo we incremented counters twice, one time for release object
and one for reloc_info.
In the main function however reloc_info references was drop much earlier than
release so keeping the pointer only on release is safe and make cleaning
process easier.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Frediano Ziglio [Wed, 3 Jun 2015 11:09:17 +0000 (12:09 +0100)]
drm/qxl: Move main reference counter to GEM object instead of TTM ones
qxl_bo structure has two reference counters, one in the GEM object and
another in the TTM object. The GEM object keep a counter to the TTM object
so when GEM counter reached zero the TTM counter (using qxl_bo_unref) was
decremented. The qxl object is fully freed (both GEM and TTM part are cleaned)
when the TTM counter reach zero.
One issue was that surface idr structure has no owning on qxl_bo objects however
it contains a pointer to qxl_bo object. This caused some nasty race condition
for instance qxl_bo object was reaped even after counter was already zero.
This patch fix these races moving main counter (the one used by qxl_bo_(un)ref)
to GEM object which cleanup routine (qxl_gem_object_free) remove the idr pointer
(using qxl_surface_evict) when the counters are still valid.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Frediano Ziglio [Wed, 3 Jun 2015 11:09:16 +0000 (12:09 +0100)]
drm/qxl: Remove format string errors
Enable format string checks for qxl_io_log and remove resulting warnings
which could lead to memory errors on different platform or just printing
wrong information.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Frediano Ziglio [Wed, 3 Jun 2015 11:09:15 +0000 (12:09 +0100)]
drm/qxl: Handle correctly failures in qxl_alloc_relase_reserved
Free resources correctly if function fails
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Frediano Ziglio [Wed, 3 Jun 2015 11:09:14 +0000 (12:09 +0100)]
drm/qxl: Fix return for qxl_release_alloc
This function return handle to allocated release object which is an int.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Frediano Ziglio [Wed, 3 Jun 2015 11:09:13 +0000 (12:09 +0100)]
drm/qxl: Handle all errors in qxl_surface_evict
Only EBUSY error was handled. This could cause code to believe
reserve was successful while it failed.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Frediano Ziglio [Wed, 3 Jun 2015 11:09:12 +0000 (12:09 +0100)]
drm/qxl: Avoid double free on error
Is we are not able to get source bo object from handle we free
destination bo object and call cleanup code however destination
object was already inserted in reloc_info array (num_relocs was
already incremented) so on cleanup we free destination again.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Frediano Ziglio [Wed, 3 Jun 2015 11:09:11 +0000 (12:09 +0100)]
drm/qxl: Fix print statement not using uninitialized variable
reloc_info[i] is not still initialized in the print statement.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Frediano Ziglio [Wed, 3 Jun 2015 11:09:10 +0000 (12:09 +0100)]
drm/qxl: Do not leak memory if qxl_release_list_add fails
If the function fails reference counter to the object is not decremented
causing leaks.
This is hard to spot as it happens only on very low memory situations.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Cc: stable@vger.kernel.org
Reviewed-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Frediano Ziglio [Wed, 3 Jun 2015 11:09:09 +0000 (12:09 +0100)]
drm/qxl: Do not cause spice-server to clean our objects
If objects are moved back from system memory to VRAM (and spice id
created again) memory is already initialized so we need to set flag
to not clear memory.
If you don't do it after a while using desktop many images turns to
black or transparents.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com>
Cc: stable@vger.kernel.org
Reviewed-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Marek Olšák [Tue, 2 Jun 2015 15:44:49 +0000 (17:44 +0200)]
drm/amdgpu: only support IBs in the buffer list (v2)
amdgpu_cs_find_mapping doesn't work without all buffers being validated,
so the TTM validation must be done first.
v2: only use amdgpu_cs_find_mapping for UVD/VCE VM emulation
Signed-off-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Ken Wang [Wed, 3 Jun 2015 13:02:01 +0000 (21:02 +0800)]
drm/amdgpu: add vram_type and vram_bit_width for interface query (v2)
Track the type of vram on the board and provide a query for it.
User mode drivers and tools want this information for determining
bandwidth information and form informational purposes.
v2: fix build when CI support is not enabled
Signed-off-by: Ken Wang <Qingqing.Wang@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Ken Wang [Thu, 4 Jun 2015 13:26:57 +0000 (21:26 +0800)]
drm/amdgpu: add ib_size/start_alignment interface query
Query the IB alignment requirements from the kernel rather
than hardcoding them in the user mode drivers.
Signed-off-by: Ken Wang <Qingqing.Wang@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Ken Wang [Wed, 3 Jun 2015 11:52:06 +0000 (19:52 +0800)]
drm/amdgpu: add me/ce/pfp_feature_version interface query
Provide this information to usermode drivers. We were previously
missing this info.
Signed-off-by: Ken Wang <Qingqing.Wang@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Ken Wang [Wed, 3 Jun 2015 09:47:54 +0000 (17:47 +0800)]
drm/amdgpu add ce_ram_size for interface query
Add a query for the CE ram size. User mode drivers
will want to use this to determine how much size
of the cache on the CE.
Signed-off-by: Ken Wang <Qingqing.Wang@amd.com>
Reviewd-by: Jammy Zhou <Jammy.Zhou@amd.com>
Ken Wang [Wed, 3 Jun 2015 09:36:54 +0000 (17:36 +0800)]
drm/amdgpu add max_memory_clock for interface query (v2)
Add a query for the max memory clock.
v2: handle the dpm enabled case properly
Signed-off-by: Ken Wang <Qingqing.Wang@amd.com>
Reviewd-by: Jammy Zhou <Jammy.Zhou@amd.com>
monk.liu [Wed, 3 Jun 2015 08:32:49 +0000 (16:32 +0800)]
drm/amdgpu: add hdp flush for gfx8 compute ring
We had forgotten to register the callback.
Signed-off-by: monk.liu <monk.liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: jammy zhou <jammy.zhou@amd.com>
monk.liu [Tue, 2 Jun 2015 07:50:33 +0000 (15:50 +0800)]
drm/amdgpu: fix no hdp flush for compute ring
No pfp on compute.
Signed-off-by: monk.liu <monk.liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Christian König [Tue, 5 May 2015 14:36:01 +0000 (16:36 +0200)]
drm/amdgpu: add HEVC/H.265 UVD support
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Christian König [Thu, 28 May 2015 20:02:26 +0000 (22:02 +0200)]
drm/amdgpu: stop loading firmware with pm.mutex locked
Loading firmware is a rather complicated process, in the end we add a
dependency between the pm mutex and the mm semaphore which results in
a harmless but annoying error message.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Christian König [Wed, 27 May 2015 08:22:47 +0000 (10:22 +0200)]
drm/amdgpu: remove mclk_lock
Not needed any more.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Mon, 11 May 2015 20:20:58 +0000 (16:20 -0400)]
drm/amdgpu: fix description of vm_size module parameter (v2)
default is 8GB, not 4GB.
v2: fix fallback setting when the user provides an invalid input
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
monk.liu [Wed, 27 May 2015 06:03:22 +0000 (14:03 +0800)]
drm/amdgpu: remove all sh mem register modification in vm flush
Leave that at the values set during init. No need to update
them repeatedly.
Signed-off-by: monk.liu <monk.liu@amd.com>
Signed-off-by: David Zhang <david1.zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <jammy.zhou@amd.com>
Marek Olšák [Wed, 27 May 2015 12:30:38 +0000 (14:30 +0200)]
drm/amdgpu: rename GEM_OP_SET_INITIAL_DOMAIN -> GEM_OP_SET_PLACEMENT
Signed-off-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
monk.liu [Tue, 26 May 2015 07:03:45 +0000 (15:03 +0800)]
drm/amdgpu: fence should be added to shared slot
Signed-off-by: monk.liu <monk.liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: jammy zhou <jammy.zhou@amd.com>
monk.liu [Tue, 26 May 2015 07:01:54 +0000 (15:01 +0800)]
drm/amdgpu: sync fence of clear_invalids (v2)
bo_va may un-initialized, fix it.
Signed-off-by: monk.liu <monk.liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <jammy.zhou@amd.com>
monk.liu [Tue, 26 May 2015 02:22:41 +0000 (10:22 +0800)]
drm/amdgpu: max_pde_used usage should be under protect
Need to take the lock when accessing this.
Signed-off-by: monk.liu <monk.liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <jammy.zhou@amd.com>
monk.liu [Mon, 25 May 2015 06:44:05 +0000 (14:44 +0800)]
drm/amdgpu: fix bug of vm_bo_map (v2)
call reservation_object_reserve_shared before amdgpu_bo_fence
Signed-off-by: monk.liu <monk.liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <jammy.zhou@amd.com>
Chunming Zhou [Fri, 24 Apr 2015 09:37:30 +0000 (17:37 +0800)]
drm/amdgpu: implement the allocation range (v3)
Pass a ttm_placement pointer to amdgpu_bo_create_restricted
add min_offset to amdgpu_bo_pin_restricted. This makes it
easier to allocate memory with address restrictions. With
this patch we can also enable 2-ended allocation again.
v2: fix rebase conflicts
v3: memset placements before using
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
yanyang1 [Fri, 22 May 2015 18:39:35 +0000 (14:39 -0400)]
drm/amdgpu: rename amdgpu_ip_funcs to amd_ip_funcs (v2)
The structure is renamed and moved to amd_shared.h to make
the component independent. This makes it easier to add
new components in the future.
v2: fix include path
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Signed-off-by: yanyang1 <young.yang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Christian König [Tue, 19 May 2015 14:08:02 +0000 (16:08 +0200)]
drm/amdgpu: drop allocation flag masks
Not needed any more.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Monk Liu <monk.liu@amd.com>
Christian König [Fri, 22 May 2015 08:55:47 +0000 (10:55 +0200)]
drm/amdgpu: drop AMDGPU_FENCE_SIGNALED_SEQ
It's causing issues with VMID handling and comparing the
fence value two times actually doesn't make handling faster.
Port of radeon commit "
d6d5c5b8364bcc4d52cddc68bcb0a330d2af20f3".
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Monk Liu <monk.liu@amd.com>
Christian König [Thu, 21 May 2015 15:03:46 +0000 (17:03 +0200)]
drm/amdgpu: port fault_reserve_notify changes from radeon
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Sonny Jiang [Tue, 12 May 2015 20:13:35 +0000 (16:13 -0400)]
drm/amdgpu: enable uvd dpm and powergating
Enable UVD dpm (dynamic power management) and powergating. UVD dpm dynamically scales the UVD
clocks on demand. Powergating turns off the power to the block when it's not in use.
Signed-off-by: Sonny Jiang <sonny.jiang@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Leo Liu [Wed, 6 May 2015 19:20:41 +0000 (15:20 -0400)]
drm/amdgpu: implement VCE two instances support
VCE 3.0 has two indentical instances in the engine, they share
the same registers name in differrent memory block distinguished
by the grbm_gfx_index, we set to master instance after init, it
will dispatch task to slave instance. These two instances will
share the same firmware, but have their own stacks and heaps.
v2: add mutex for using grbm_gfx_index
Signed-off-by: Leo Liu <leo.liu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Leo Liu [Wed, 6 May 2015 18:31:27 +0000 (14:31 -0400)]
drm/amdgpu: recalculate VCE firmware BO size
Firmware required BO size changes in terms of ASIC family
Signed-off-by: Leo Liu <leo.liu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Alex Deucher [Thu, 21 May 2015 06:24:48 +0000 (02:24 -0400)]
drm/amdgpu: remove unused TRACE_SYSTEM_STRING define
Port of
77cb2fea1e5fc4b083dd967f231bbf6edd96150e to
amdgpu.
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Marek Olšák [Thu, 14 May 2015 21:48:26 +0000 (23:48 +0200)]
drm/amdgpu: rework tiling flags
Signed-off-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Marek Olšák [Thu, 14 May 2015 21:03:57 +0000 (23:03 +0200)]
drm/amdgpu: don't set unused tiling flags
Signed-off-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Christian König [Mon, 18 May 2015 14:05:57 +0000 (16:05 +0200)]
drm/amdgpu: actually use the VM map parameters
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Christian König [Mon, 18 May 2015 12:37:27 +0000 (14:37 +0200)]
drm/amdgpu: validate amdgpu_vm_bo_map parameters
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Christian König [Wed, 13 May 2015 12:30:53 +0000 (14:30 +0200)]
drm/amdgpu: enforce AMDGPU_GEM_CREATE_NO_CPU_ACCESS
Deny user and kernel mapping if we said we never want to do so.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Christian König [Wed, 13 May 2015 12:21:06 +0000 (14:21 +0200)]
drm/amdgpu: fix VM_CONTEXT*_PAGE_TABLE_END_ADDR handling
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Mon, 18 May 2015 22:15:07 +0000 (18:15 -0400)]
drm/amdgpu: retry dcpd fetch
Retry the dpcd fetch several times. Some eDP panels
fail several times before the fetch is successful.
bug:
https://bugs.freedesktop.org/show_bug.cgi?id=73530
Ported from radeon.
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Mon, 18 May 2015 22:12:02 +0000 (18:12 -0400)]
drm/amdgpu: simplify DPCD debug output
Use %*ph rather than walking the array.
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Mon, 18 May 2015 22:09:23 +0000 (18:09 -0400)]
drm/amdgpu: make some DP parameters const
Ported from similar radeon patch.
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Fri, 15 May 2015 15:52:18 +0000 (11:52 -0400)]
drm/amdgpu: take the mode_config mutex when handling hpds
Since we may modify display state.
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Marek Olšák [Tue, 5 May 2015 19:13:49 +0000 (21:13 +0200)]
drm/amdgpu: add and implement the GPU reset status query
Signed-off-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Alex Deucher [Tue, 12 May 2015 17:10:05 +0000 (13:10 -0400)]
drm/amdgpu: add some new tonga pci ids
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Tue, 12 May 2015 17:06:45 +0000 (13:06 -0400)]
drm/amdgpu: add new bonaire pci id
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Jammy Zhou [Wed, 13 May 2015 14:52:42 +0000 (22:52 +0800)]
drm/amdgpu: rewording some left radeons
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Jammy Zhou [Wed, 13 May 2015 14:49:04 +0000 (22:49 +0800)]
drm/amdgpu: switch to amdgpu folder for firmware files v2
v2: keep using radeon folder for CIK
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Jammy Zhou [Tue, 12 May 2015 15:17:19 +0000 (23:17 +0800)]
drm/amdgpu: do necessary NULL check
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Jammy Zhou [Tue, 12 May 2015 14:46:45 +0000 (22:46 +0800)]
drm/amdgpu: expose the max virtual address
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Christian König [Mon, 11 May 2015 13:34:59 +0000 (15:34 +0200)]
drm/amdgpu: fix context switch
Properly protect the state and also handle submission failures.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Monk Liu <monk.liu@amd.com>
Christian König [Mon, 11 May 2015 12:32:17 +0000 (14:32 +0200)]
drm/amdgpu: fix dereference before check
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Monk Liu <monk.liu@amd.com>
Christian König [Mon, 11 May 2015 12:10:34 +0000 (14:10 +0200)]
drm/amdgpu: cleanup HDP flush handling
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Monk Liu <monk.liu@amd.com>
Christian König [Mon, 11 May 2015 11:52:09 +0000 (13:52 +0200)]
drm/amdgpu: always emit GDS switch
Otherwise a process can access the GDS data of another process.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Monk Liu <monk.liu@amd.com>
Jammy Zhou [Mon, 11 May 2015 15:49:34 +0000 (23:49 +0800)]
drm/amdgpu: add CE preamble flag v3
The CE preamble IB can be dropped for the same context
v2: use the flags directly
v3: remove 'CE' for potential preamble usage by other rings
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Jammy Zhou [Mon, 11 May 2015 15:41:41 +0000 (23:41 +0800)]
drm/amdgpu: add flags for amdgpu_ib structure
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Jammy Zhou [Fri, 8 May 2015 14:18:47 +0000 (22:18 +0800)]
drm/amdgpu: check context id for context switching (v2)
check the filp is not robust, and sometimes different contexts may
have same filp value.
v2: check both filp and ctx_id
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Jammy Zhou [Fri, 8 May 2015 09:29:40 +0000 (17:29 +0800)]
drm/amdgpu: add ctx_id to the WAIT_CS IOCTL (v4)
It is required to support fence per context.
v2: add amdgpu_ctx_get/put
v3: improve get/put
v4: squash hlock fix
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Jack Xiao [Fri, 8 May 2015 06:46:49 +0000 (14:46 +0800)]
drm/amdgpu: allow unaligned memory access (v2)
Set up the CP and SDMA for proper unaligned memory access.
Required for OpenCL 2.x
v2: udpate commit message
Signed-off-by: Jack Xiao <Jack.Xiao@amd.com>
Reviewed-by: Monk Liu <monk.liu@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Marek Olšák [Tue, 5 May 2015 18:52:00 +0000 (20:52 +0200)]
drm/amdgpu: make the CTX ioctl thread-safe
The existing locks were protecting the list, but not the elements.
v2: rename hlock to lock
Signed-off-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Marek Olšák [Mon, 4 May 2015 22:56:45 +0000 (00:56 +0200)]
drm/amdgpu: remove unsafe context releasing
If ctx was released between put and get, then "get" would crash.
Signed-off-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Christian König [Mon, 4 May 2015 11:20:36 +0000 (13:20 +0200)]
drm/amdgpu: fix userptr lockup
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Monk Liu <monk.liu@amd.com>
monk.liu [Thu, 7 May 2015 18:19:18 +0000 (14:19 -0400)]
drm/amdgpu: fix userptr BO unpin bug (v2)
sg could point to array of contigiouse page*, only free page could lead
to memory leak.
v2: use iterator
Signed-off-by: monk.liu <monk.liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>